From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Burakov, Anatoly" <anatoly.burakov@intel.com>
Subject: Re: [PATCH] eal/vfio: fix sPAPR IOMMU mapping
Date: Tue, 7 Aug 2018 09:39:36 +0100
Message-ID: <38b342bb-3f41-7912-83d5-a9f207032cd2@intel.com>
References: <20180806084309.5489-1-tyos@jp.ibm.com> <4068541.lKVHRX04u0@xps>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
Cc: dev@dpdk.org, Takeshi Yoshimura <tyos@jp.ibm.com>,
 Takeshi Yoshimura <t.yoshimura8869@gmail.com>
To: Thomas Monjalon <thomas@monjalon.net>, chaozhu@linux.vnet.ibm.com
Return-path: <dev-bounces@dpdk.org>
Received: from mga04.intel.com (mga04.intel.com [192.55.52.120])
 by dpdk.org (Postfix) with ESMTP id 0C56C1B068
 for <dev@dpdk.org>; Tue,  7 Aug 2018 10:39:41 +0200 (CEST)
In-Reply-To: <4068541.lKVHRX04u0@xps>
Content-Language: en-US
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <https://mails.dpdk.org/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://mails.dpdk.org/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <https://mails.dpdk.org/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

On 06-Aug-18 10:50 AM, Thomas Monjalon wrote:
> 06/08/2018 10:43, Takeshi Yoshimura:
>> Commit 73a639085938 ("vfio: allow to map other memory regions")
>> introduced a bug in sPAPR IOMMU mapping. The commit removed necessary
>> ioctl with VFIO_IOMMU_SPAPR_REGISTER_MEMORY. Also, vfio_spapr_map_walk
>> should call vfio_spapr_dma_do_map instead of vfio_spapr_dma_mem_map.
> 
> This is fixing an old patch:
> 	http://git.dpdk.org/dpdk/commit/?id=73a639085938
> 
>> Fixes: 73a639085938 ("vfio: allow to map other memory regions")
> 
> You probably want it to be backported in previous release,
> so you need to add Cc: stable@dpdk.org
> 
>> Signed-off-by: Takeshi Yoshimura <tyos@jp.ibm.com>
> 
> It is common to have bugs in sPAPR implementation.
> How can we be sure this one is OK?
> Should it be added in last minute of 18.08?
> Or can it wait 18.11?
> 
> Adding Anatoly and Chao who are maintainers to decide.
> 

The patch appears to be correct - we did have a 
VFIO_IOMMU_SPAPR_REGISTER_MEMORY ioctl in the old map() function, but 
not in the new one.

However, i agree with Thomas - without more testing from other sPAPR 
users/IBM, i would be hesitant to allow it in at the last minute. In any 
case, this code has been there for a while and no one has reported any 
issues - so this can probably wait until 18.11, seeing how this codepath 
is so popular :)

-- 
Thanks,
Anatoly